November 7th, 2009This release offers a fix to a long standing issue in ITK regarding the computation of physical coordinates associated with pixels. This fix is enabled by default, but if you need to revert it to the previous behavior for backward compatibility reasons, you can disable it by turning off the CMake flag:
· ITK_USE_CENTERED_PIXEL_COORDINATES_CONSISTENTLY. Thanks to Tom Vercauteren and Michel Audette for their hard work on getting this difficult issue fixed.
· Many improvements and fixes in the I/O infrastructure were contributed by Brad Lowekamp, making it now possible to manage large image files through the streaming infrastructure of the data pipeline. In particular Brad contributed examples illustrating how to process the entire Visible Human dataset via an ITK pipeline.
July 21st, 2009· Additional color map functors. Based on an Insight Journal article by N. Tustison, H. Zhang, G. Lehmann, P. Yushkevich, and J. Gee, these classes provide a framework for converting intensity-valued images to user-defined RGB colormap images. Along with the framework, the classes also include several colormaps that can be readily applied for visualization of images, or adapted to generate user desired colormaps. Details are available from "Meeting Andy Warhol Somewhere Over the Rainbow: RGB Colormapping and ITK", which can be read in the January Source or on the Insight Journal (http://hdl.handle.net/1926/1452).
· Direct FFT-based sinogram to image tomographic reconstructions. Based on an Insight Journal article by D. Zosso, Cuadra M Bach, and J Thiran, these classes provide the implementation of a direct Fourier method for tomographic reconstruction, applicable to parallel-beam x-ray images. Details are available from "Direct Fourier Tomographic Reconstruction Image-To-Image Filter" (http://hdl.handle.net/1926/585).
· Efficient label map operations. Based on an Insight Journal article by G. Lehmann, the added classes are the initial components of a 70+ class label map morphology module. These classes provide for the efficient representation of label maps and for conversion between current ITK label images and the efficient storage format. Details are available from "Label Object Representation and Manipulation with ITK", which can be read in the January Source or on the Insight Journal (http://hdl.handle.net/1926/584).
· Fractal image dimension calculations. Based on an Insight Journal article by N. Tustison and J. Gee, this classes provides an algorithm for converting a scalar image to a fractal dimension image. Details are available from "Stochastic Fractal Dimension Image" (http://hdl.handle.net/1926/1525).
· Additional level set algorithms including support for multi-component level sets. Based on an Insight Journal article by K. Mosaliganti, B. Smith, A. Gelas, A. Gouaillard and S. Megason, these classes model an image as a constant intensity background with constant intensity foreground components. Segmentation results from a previous time point can be leveraged in processing a new time point, and real time tracking for fluoresence microscope applications can be attained. Details are available from "Cell Tracking using Coupled Active Surfaces for Nuclei and Membranes" (http://hdl.handle.net/10380/3055).
· Helper class for initializing BSpline Transforms. Based on an Insight Journal Article by L. Ibanez, M. Turek, S. Aylward, and M. Audette, this class allows a BSpline transform to be easily initialized to reasonable values prior to a registration operation. Details are available in "Helper class for initializing the grid parameters of a BSpline deformable transform by using an image as reference" (http://hdl.handle.net/1926/1338).
Find us on Google+
